Hugging Face's logo

Debmalya
/
ComcastClassifier

Model card Files Community
ComcastClassifier
File size: 4,829 Bytes 
c9665af
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
 

import streamlit as st 
import joblib,os
import spacy
import pandas as pd
nlp = spacy.load("en_core_web_sm")
import matplotlib.pyplot as plt 
import matplotlib
matplotlib.use("Agg")
from wordcloud import WordCloud


# load Vectorizer 
complaints_vectorizer = open("models/tfidf_vect.pickle","rb")
complaints_cv = joblib.load(complaints_vectorizer)

def load_prediction_models(model_file):

	loaded_model = joblib.load(open(os.path.join(model_file),"rb"))
	return loaded_model

# Get the Keys
def get_key(val,my_dict):
	for key,value in my_dict.items():
		if val == value:
			return key


def main():

	"""Telecom Complaints Classifier"""
	st.title("Telecom Complaints - Classification App")
	
	# Layout Templates
	html_temp = """
	<div style="background-color:#464e5f;padding:10px;border-radius:10px;margin:10px;">
	<h1 style="color:white;text-align:center;"> ML - Telecom Complaints Classifier </h1>
	<img src="https://www.w3schools.com/howto/img_avatar.png" alt="Avatar" style="vertical-align: middle;width: 50px;height: 50px;border-radius: 50%;" >
	<p style="text-align:justify">{}</p>
	</div>
	"""
	title_temp ="""
	<div style="background-color:#464e5f;padding:10px;border-radius:10px;margin:10px;">
	<h4 style="color:white;text-align:center;">{}</h1>
	<img src="https://www.w3schools.com/howto/img_avatar.png" alt="Avatar" style="vertical-align: middle;float:left;width: 50px;height: 50px;border-radius: 50%;" >
	<h6>Author:{}</h6>
	<br/>
	<br/> 
	<p style="text-align:justify">{}</p>
	</div>
	"""
	article_temp ="""
	<div style="background-color:#464e5f;padding:10px;border-radius:5px;margin:10px;">
	<h4 style="color:white;text-align:center;">{}</h1>
	<h6>Author:{}</h6> 
	<h6>Post Date: {}</h6>
	<img src="https://www.w3schools.com/howto/img_avatar.png" alt="Avatar" style="vertical-align: middle;width: 50px;height: 50px;border-radius: 50%;" >
	<br/>
	<br/>
	<p style="text-align:justify">{}</p>
	</div>
	"""

	
	st.markdown(html_temp,unsafe_allow_html=True)

	activity = ['Prediction','NLP','About']
	choice = st.sidebar.selectbox("Select Activity",activity)


	if choice == 'Prediction':
		st.info("Prediction with ML")
		complaints_text = st.text_area("Enter Complaints Here","Type Here")
		all_ml_models = ["Decision Tree", "GradientBoost"]
		model_choice = st.selectbox("Select Model",all_ml_models)

		prediction_labels = {'Closed': 0, 'Open': 1, 'Pending': 2, 'Solved': 3}
		if st.button("Classify"):
			st.text("Original Text:\n{}".format(complaints_text))
			vect_text = complaints_cv.transform([complaints_text]).toarray()
			if model_choice == 'Decision Tree':
				predictor = load_prediction_models("models/dtcpred.pickle")
				prediction = predictor.predict(vect_text)
				# st.write(prediction)
			elif model_choice == 'GradientBoost':
				predictor = load_prediction_models("models/gbcpred.pickle")
				prediction = predictor.predict(vect_text)
				# st.write(prediction)
                
                
			final_result = get_key(prediction,prediction_labels)
			st.success("Complaints Categorized as: {}".format(final_result))

	elif choice == 'NLP':
		st.info("Natural Language Processing of Text")
		raw_text = st.text_area("Enter Customer Complaints Here","Type Here")
		nlp_task = ["Tokenization","Lemmatization","Named Entity Recognition(NER)","Parts-of-Speech(POS) Tags"]
		task_choice = st.selectbox("Choose NLP Task",nlp_task)
		if st.button("Analyze"):
			st.info("Original Text:\n{}".format(raw_text))

			docx = nlp(raw_text)
			if task_choice == 'Tokenization':
				result = [token.text for token in docx ]
			elif task_choice == 'Lemmatization':
				result = ["'Token':{},'Lemma':{}".format(token.text,token.lemma_) for token in docx]
			elif task_choice == 'Named Entity Recognition(NER)':
				result = [(entity.text,entity.label_)for entity in docx.ents]
			elif task_choice == 'Parts-of-Speech(POS) Tags':
				result = ["'Token':{},'POS':{},'Dependency':{}".format(word.text,word.tag_,word.dep_) for word in docx]

			st.json(result)

		if st.button("Tabulize"):
			docx = nlp(raw_text)
			c_tokens = [token.text for token in docx ]
			c_lemma = [token.lemma_ for token in docx ]
			c_pos = [token.pos_ for token in docx ]

			new_df = pd.DataFrame(zip(c_tokens,c_lemma,c_pos),columns=['Tokens','Lemma','POS'])
			st.dataframe(new_df)


		if st.checkbox("WordCloud"):
			c_text = raw_text
			wordcloud = WordCloud().generate(c_text)
			plt.imshow(wordcloud,interpolation='bilinear')
			plt.axis("off")
			st.set_option('deprecation.showPyplotGlobalUse', False)
			st.pyplot()

	else:
		st.write("")
		st.subheader("About")
		st.write("")
	
		st.markdown("""
        ### NLP Complaints Classifier With Different Models (With Streamlit)
        Python Tools Used: spacy, pandas, matplotlib, wordcloud, Pillow(PIL), Joblib
        """)


if __name__ == '__main__':
	main()